Residual Stress After Bending Sheetmetal

[Zach125]

First post on Eng-Tips!

This is as much a question as it is something I'd like to get a discussion going on. I'll phrase it like a question:

When bending sheetmetal, the material at the bend has clearly been stressed past the yield stress but less than the ultimate/failure stress. So there are residual stresses at the bend and material has yielded before the part is ever loaded.

When analyzing a part for structural integrity, is it appropriate/prudent/good practice to consider this residual stress/yielding? Or is it ignored to make the problem more manageable? Does anyone have experience either way?

 

[rb1957]

well, there are two ways of doing what you talk about ...

1) plastically form the s/m (usually in O condition) and then heat treat to remove the strains and to increase the temper (= strength) of the finished part. Thus no residual stresses.

2) plastically form the s/m and live with the residual stresses. Forming to standard shop practices (eg 3t bend rad) has shown (over the 50+ years we've been doing this) that this ok.

Quando Omni Flunkus Moritati

 

[Zach125]

rb1957, thanks for weighing in!

Option 1 would certainly be the best structural solution. Unfortunately, the specific design I'm working drives us to Option 2 for various reasons (cost, material availability...).

The reason I'm a bit concerned is that the bend radius is the tightest within the recommended forming range and the part will be loaded in the exact same direction as when it was bent. Thus it seems unsafe to neglect the residual stress.

I'm wondering if there is any standard/textbook way of analyzing this problem, short of some complicated non-linear FEA modeling the bending process and capturing whatever residual stresses develop during forming.

 

[rb1957]

build in plenty of margin ...

Quando Omni Flunkus Moritati

 

[Sparweb]

This has been analyzed and tested in the past. Back in the 50's I'm sure Lockheed had some pretty fancy FEM software packages to work with, too.
Teasing aside, the answer to your question is "yes" and you will find much detail in the textbooks under this forum subject's FAQ. here:

http://www.eng-tips.com/faqs.cfm?fid=919

I believe some research under "tension clips", either with the forum search tool, or the old-fashioned google approach, may also reveal some design charts that you can use, but the source may be uncertain.

STF

 

[Zach125]

The Bruhn book has some semi-applicable nuggets (chapts. A13, C3), but I couldn't find anything directly addressing my question. No design charts either, as far as I can tell.

My Google efforts have turned up some interesting stuff on FEA modeling of springback after bending, but nothing from a stress analysis perspective.

For what it's worth, in my experience I've done as rb1957 suggested, either 1) heat treating the part to remove the residual stress or 2) bending per standard practice, leaving a solid design margin, and not worrying about it. I also spoke to a structural DER and he echoed the same sentiments that option 2 has been historically shown to work and be acceptable.

But it's my job to worry about this kind of thing, so I posed the question. I'm hoping to draw on some collective wisdom on the subject. SparWeb, if you know of any specific design charts or applicable references, I would be indebted to you if you could pass them along.

Perhaps this question would be more suited for a mechanical or materials forum?

 

[rb1957]

i worried about one of our forming processes ... "cinching" to form an arc (for a fuselage shear cleat).

we did a bunch of FEA ... it showed the most unexpected results, i think it showed a stress peak on the free edge of the undeformed flange of the angle (the one to attach to the fuselage), i was expecting to see nasty stresses at the corner. it fell onto the back burner but it'll have to be looked at again (i thought we used plastic FEA (we should have), maybe we'll try photo-stress ?)

residual stresses are something to worry about; saying we've done it that way for 50+ years could easily be the beginning of a "mayday" episode.

Quando Omni Flunkus Moritati

 

[Zach125]

I found a good study of this topic. See page 10, figure 12 for what I'm worried about. A bit concerning to be loading a part that already has residual stress.

Link

I've seen statements that the hysteresis loop returns the stress to zero when the sheet is unloaded after being bent, just with permanent deformation, but I don't buy that.

It seems like there ought to be design tables for this somewhere. I've been in the industry a couple years and haven't seen any.

 

[rb1957]

"I've seen statements that the hysteresis loop returns the stress to zero when the sheet is unloaded after being bent, just with permanent deformation, but I don't buy that." i'm with you on that ! if the thing is bent, then it has permanent deformation (= internal stresses). they're only removed by heat treating, (aging, maybe) and not by wishing them away !

Quando Omni Flunkus Moritati

 

[ornerynorsk]

Engineering Material and Metallurgy has a brief synopsis of this under "Cold Forming".

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[Zach125]

I played around a bit with modeling the bending process and then applying load to the part. There is still a decent margin, but it was not insignificantly reduced due to the residual stresses. So it definitely matters, can't confidently say exactly how much since I just took a quick pass at it.

Long story short, best to just heat treat it if it's a structural part and save yourself the headache and worry.